diff options
| author |   David S. Miller <davem@davemloft.net> | 2016-06-10 23:58:29 -0700 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2016-06-10 23:58:29 -0700 |
| commit | e69f73bfecb0178ae6bd20eb778211739cd71fab (patch) | |
| tree | 05d9f30419d6a5ef9b503487661959ef29c2fec5 /net/ipv6/ping.c | |
| parent | Merge branch 'mdio-iProc-SOC' (diff) | |
| parent | net_sched: remove generic throttled management (diff) | |
| download | linux-dev-e69f73bfecb0178ae6bd20eb778211739cd71fab.tar.xz linux-dev-e69f73bfecb0178ae6bd20eb778211739cd71fab.zip | |
Merge branch 'remove-qdisc-throttle'
Eric Dumazet says:
====================
net_sched: remove qdisc_is_throttled()
HTB, CBQ and HFSC pay a very high cost updating the qdisc 'throttled'
status that nothing but CBQ seems to use.
CBQ usage is flaky anyway, since no qdisc ->enqueue() updates the
'throttled' qdisc status.
This looks like some 'optimization' that actually cost more than code
without the optimization, and might cause latency issues with CBQ.
In my tests, I could achieve a 8 % performance increase in TCP_RR
workload through HTB qdisc, in presence of throttled classes,
and 5 % without throttled classes.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to '')
0 files changed, 0 insertions, 0 deletions